Methods of Research | Psychology for UPSC Optional (Notes) PDF Download

Non-Experimental Methods

1. Naturalistic Observation

• In certain research scenarios, it is important for researchers to simply observe and understand the behaviors of animals or individuals. The most effective approach to studying their behavior is through naturalistic observation, where scientists observe behavior in real-world settings without attempting to manipulate or control the situation. Naturalistic observation takes place in various locations such as homes, day-care centers, and other natural environments. For instance, if researchers wanted to understand how adolescents interact with the opposite sex in social settings, they might observe their behavior at a mall on a weekend night.
• The primary advantage of naturalistic observation is that it provides researchers with a genuine understanding of how behavior naturally occurs by directly observing it. However, it is important to note that individuals or animals who are aware of being observed may alter their behavior, known as the observer effect. To mitigate this, observers often remain concealed or become participants in the group being studied, a technique known as participant observation.
• One significant drawback of naturalistic observation is the potential for observer bias, which occurs when the observer has preconceived expectations or biases about the behavior they anticipate seeing. This may lead to selective observation, where the observer only notices actions that align with their expectations while disregarding those that do not. Additionally, each naturalistic setting is unique and unlike any other, making it challenging to generalize observations across different settings. Even if the setting appears similar, conditions may vary from one observation to another as researchers lack complete control over the natural world.

2. Archival Research

• In this method, researchers do not directly gather data but instead acquire it from public records, archives, and similar sources. They analyze the obtained data to draw conclusions. This approach can be valuable in various ways, such as when collecting data on suicides and homicides, where alternative methods may not be available. Archival data refers to existing records or archives that the researcher examines or selects for analysis. Archival research is often utilized when conducting an experiment related to the variables of interest is impractical due to logistical or ethical reasons.
• However, archival research has its limitations. Firstly, most archival data are collected for non-research purposes, such as governmental or private agency records, and may not perfectly align with the researcher's needs. Secondly, since archival research is conducted retrospectively, it can be challenging to rule out alternative hypotheses or determine causality for observed correlations. Researchers relying on archival data are susceptible to biases that might have influenced the data collection process. Notably, police records are known to be subject to bias, and many types of crime often go unreported to the authorities.

3. Content Analysis

• Content analysis, also referred to as document analysis, is a systematic method used to examine communications or existing records and documents. Instead of directly questioning respondents or observing their behavior, the content analyst analyzes the communications or documents generated by the respondents to determine the frequency or proportion of certain elements. 
• Content or document analysis relies on various primary sources of data, including letters, autobiographies, diaries, reports, printed forms, academic works such as themes or compositions, books, periodicals, bulletins, catalogues, syllabi, court decisions, pictures, films, and cartoons. It is the responsibility of the researchers to establish the reliability and validity of the data selected for analysis. Content analysis can also be applied to responses from projective tests, various forms of verbal materials, and materials specifically created for research purposes.

Merits and Demerits

• Firstly, content analysis can be applied to a diverse range of materials, including creativity, attitudes, ethnocentrism, stereotypes, curriculum changes, values, interests, religiosity, college budgets, and more. 
• Secondly, content analysis can be utilized to investigate the impact of experimental manipulations on dependent variables. For instance, when studying the effect of practice on children's handwriting improvement, content analysis holds equal importance to any experimental design.
• Thirdly, content analysis is valuable for validating other observation methods. For example, in validating a self-disclosure inventory, individuals may be reluctant to provide personal information directly. However, by posing projective-type questions and analyzing the content of their responses, the inventory can be validated against the content analysis.
• Despite these merits, it is essential to exercise caution when employing content analysis due to the complexities involved.

Surveys

• Survey methods are widely employed for gathering scientific information through the collection of data by asking questions and recording people's responses. These methods serve various purposes, with one common objective being the estimation of population characteristics. For instance, a survey may aim to determine the percentage of individuals who hold supportive or opposing views on specific social issues, such as the provision of job reservations for women. Census data and public opinion polls conducted by different agencies serve as notable examples of surveys.
• Surveys are also useful for testing hypotheses regarding relationships between variables. Researchers may investigate the impact of certain events on people's behavior, such as conducting surveys after an earthquake in Bhuj, Gujarat, to assess the earthquake's effect on individuals' lives. During the survey process, researchers define the study population and draw a representative sample from it. Various sampling procedures are employed, including random sampling, where every population member has an equal and independent chance of being included in the sample.
• Stratified random sampling is often utilized, which involves representing two or more subgroups in proportions predetermined by their presence in the population. In some cases, groups are selected through cluster sampling, where clusters or groupings are chosen from a larger population. The sample size is also determined, as the ability to generalize findings depends on the sample size used in the survey.
• Survey methods can be classified into different categories based on the means of data collection, such as personal interviews, mail questionnaires, telephone surveys, internet surveys, web surveys, and more.

Advantages

• Survey methods offer a wide-ranging capacity, allowing researchers to gather a substantial amount of information by studying a larger population. This method is known for its heightened accuracy, as highlighted by Kerlinger (1986) who stated that the accuracy of properly conducted samples often surprises even experts in the field. A sample size of 600 to 700 individuals or families can provide a remarkably precise representation of a community, encompassing its values, attitudes, and beliefs.
• Survey methods have been extensively utilized across almost all social sciences, indicating their interdisciplinary significance. Consequently, such research endeavors serve as valuable raw materials for the expanding realm of "gross disciplinary research" (Campbell & Katona, 1953).
• The survey method is widely recognized as a vital and indispensable tool for studying social attitudes, beliefs, values, and other related aspects with a cost-effective level of accuracy.

Disadvantages

• Survey methods tend to remain on the surface and lack the ability to delve deep into the underlying issues being investigated.
• Conducting surveys can be time-consuming and require a significant amount of financial investment.
• Despite the overall accuracy of survey research, it is still susceptible to sampling errors. There is always a probability, typically one in twenty or one in a hundred, of an error occurring that is more severe than minor fluctuations, potentially distorting the validity of the obtained results.
• The survey method necessitates expertise, research knowledge, and a certain level of sophistication on the part of the researcher. In other words, the researcher must possess knowledge of sampling techniques, questionnaire construction, interviewing skills, and data analysis.

Field Studies

• Field studies are scientific inquiries conducted after the fact, aiming to uncover relationships and interactions among sociological, psychological, and educational variables within real social structures. These studies systematically explore relationships and test hypotheses that are formulated based on real-life situations. The investigator in a field study observes the social or institutional context and examines the relationships among the attitudes, values, perceptions, and behaviors of individuals and groups within that context. Typically, no independent variables are manipulated in field studies.
• Katz (1953) has categorized field studies into two broad types: exploratory and hypothesis testing. Exploratory field studies focus on understanding the existing conditions rather than predicting relationships. They serve three purposes: (1) identifying significant variables within the field situation, (2) uncovering relationships among variables, and (3) establishing a foundation for later, more systematic and rigorous hypothesis testing. It is important to acknowledge that preliminary methodological and measurement investigations are often necessary before hypothesis testing can take place in scientific research.
• The second subtype of exploratory field studies, aimed at discovering or uncovering relationships, is essential for advancing scientific knowledge in the social sciences. Field studies exhibit strengths in realism, the significance of variables, theoretical orientation, and heuristic value. Realism is inherent in field studies, and they offer rich opportunities for discovery. However, researchers must exercise caution to remain focused on their specific research problem as they may encounter numerous intriguing ideas along the way. Field studies have a strong heuristic quality, prompting the researcher to explore new avenues of investigation.
• Despite these strengths, field studies possess scientific weaknesses compared to laboratory experiments. Their most significant weakness lies in their ex-post facto nature. Additionally, field studies often lack precision in measuring variables within the real-world context. Practical challenges such as feasibility, cost, sampling, and time further contribute to the limitations of field studies. Therefore, field researchers must possess skills beyond investigation, acting as salespeople, administrators, and entrepreneurs.

Case Study

• The case study is a significant form of non-experimental research that involves organizing social data to gain insights into social reality. It aims to maintain the integrity of the social object being studied and examines it as a cohesive unit. The unit of analysis in a case study can range from an individual, a family, a social group, a social institution, or even a community (Goode & Hatt, 1981; Best & Kahn, 1992). A case study may utilize various methods such as interviews, observations, and psychological tests. It is particularly valuable in fields like clinical psychology and human development, offering a detailed examination of a single individual.
• Through the use of a case study, researchers can gain an in-depth understanding of unique aspects of a person's life that cannot be replicated due to practical or ethical constraints. It allows researchers to explore a person's fantasies, hopes, fears, traumatic experiences, upbringing, and other factors that contribute to their mind and behavior. Case studies provide a narrative or detailed description of events occurring in a person's life. Notably, Freud's development of psychoanalytic theory was influenced by his observations and reflections on individual cases.
• However, it is important to exercise caution when generalizing from individual cases as they are akin to naturalistic observations. Case studies provide detailed and in-depth portrayals of individuals' lives, but the reliability of our judgments regarding a particular case is unknown. The issue of validity in a single case study is a serious concern. Therefore, it is recommended that researchers employ objective measurement techniques, multiple sources of information, and frequent assessments of relevant variables. The selection of cases must be done carefully to ensure they represent the variable under investigation, and researchers should have sufficient access to the chosen cases. Thorough planning of data collection is crucial, and throughout the process, investigators must maintain a chain of evidence that links the various data sources to address the research questions effectively.

Comparison between Research Methodologies

The Experiment

• The concept of experimentation is not commonly utilized in the field of public administration and management, although certain experimental elements such as simulations, gaming, and policy evaluation are occasionally employed. Van Thiel (2014) identifies various types of experiments, including the classic experiment, simulation/gaming, and field experiment. While experiments offer high reliability due to the level of control they provide, they often lack realism as controlled experiments differ significantly from real-life situations. In a classic experiment, subjects are typically individuals who are divided into an experimental group and a control group. 
• The experimental group is exposed to a specific stimulus, while the control group is not. These experiments are time-based, and once the experiment concludes, the effects of the stimulus can be analyzed. Usually conducted in a controlled space or area, the researcher limits the number of variables that may influence the experiment. The minimum set of variables in an experiment consists of an independent variable (stimulus) and a dependent variable (the outcome being measured). Another type of experiment is the field experiment, which takes place outside of controlled spaces or laboratories. In the realm of public administration, field experiments are typically conducted by government or public entities rather than individual researchers. Experiments can also take the form of simulations and gaming, where reality is imitated within a controlled setting manipulated by the researcher. Unlike classic experiments, gaming and simulations do not involve dividing groups into control and experimental groups, and multiple variables can be introduced into the experiment.

The Survey

• The survey is perhaps one of the most well-known and widely used research methods. It allows for the collection of a significant amount of information from thousands of people and can be tailored to various research questions, polls, or opinions. A standard survey is typically large-scale, involves multiple variables, and targets numerous respondents (Van Thiel, 2014). To facilitate large-scale data collection, researchers employ standardized forms of measurement, such as answer scales or numerical answer categories (DeVellis, 2012). The most common type of survey is a written questionnaire containing closed-ended questions with fixed answer options (Van Thiel, 2014). 
• For example, respondents may be asked to rate how safe they feel in the event of a hurricane impacting their city, with answer choices ranging from "very safe" to "very unsafe." The process of a written questionnaire involves five key steps: questionnaire design, pilot testing, determining the respondent sample size or type, administering the questionnaire, and entering the respondent data into a database for extraction and analysis. Attention must be given to ensure the questionnaire is clear, avoids leading statements, utilizes consistent answer categories, and covers a comprehensive range of options. 
• The Likert scale is often employed in surveys as it generates accurate and consistent statistical data. Additionally, the layout and organization of the survey or questionnaire are crucial. It should begin with an introduction or instructions outlining the purpose of the study and the identity of the researcher. The order of the questions should follow a logical and easy-to-follow structure, and the layout should be visually appealing and clear to respondents.

Case Study

• A case study is a research strategy where one or more cases within a real-life setting are observed. A case can encompass various subjects, such as a law, process, group, city, country, or relationship. Typically, a case study adopts a holistic approach, gathering qualitative data on all aspects related to the case. Case studies can be inductive or deductive and often yield results with high validity and reliability. They are conducted in real-life scenarios (field experiments) and are usually designed to address or find solutions to specific issues or problems. Unlike surveys, case studies tend to focus on a limited number of situations but provide more detailed information, aiming for depth rather than breadth (Timney Bailey, 1992). 
• Several key elements should be considered when conducting case studies, such as the number of cases (single, multiple, contrasting, homogeneous), the number of measurements (time frame, period, spacing), and the research methods to be employed. To enhance the validity and reliability of a case study, researchers often employ triangulation or mixed methods, combining techniques such as observation, content analysis, and interviews. Following a proper case study protocol helps researchers stay on track and plan their day-to-day tasks and activities. Case study protocols encompass procedures, methods, and analysis techniques.